Graph‐based local climate classification in Iran

In this paper, we introduce a novel graph-based method to classify the regions with similar climate in local area. We refer our proposed as graph partition based (GPBM). Our attempts overcome shortcomings of current state-of-the-art methods literature. It has no limit on number variables that can be used and also preserves nature data. To illustrate capability algorithm, benchmark its performance other classification techniques. The data are collected from 24 synoptic stations Fars province southern Iran. include seven stored time series 1951 2017. results exhibit performs more realistic less computational time. save information during process is therefore efficient further analysis. Furthermore, using method, seasonal graphs better investigate changes. best knowledge, first system.